Sunil GillApp marketing is everything that gets an app found, installed, used and paid for. Most of the advice written about it is a list of channels: do ASO, do social, do influencers, run ads. The list is not the hard part. The order is.
Apps fail at marketing by doing things in the wrong sequence, usually by paying for installs before the app keeps anyone. This guide is the order I would follow for a new app on iOS and Android, with the numbers that tell you when to move to the next step.
The order that works
- Make the store listing convert.
- Launch small and fix the first session.
- Build the free channels: store search and Google.
- Work out what an install is worth.
- Only then buy installs, one channel at a time.
Each step makes the next one cheaper. A listing that converts 12% of visitors instead of 6% halves the cost of every install you later pay for by the tap. An app that keeps its users earns more from each one, which raises what you can afford to bid. Skip ahead and you pay full price for a leaky funnel.
Step 1: make the store listing convert
Every channel ends at your product page, so it comes first. It is also where most installs start. Apple says 70% of App Store visitors use search to find apps and almost 65% of downloads happen directly after a search.
- Name your app "Brand: main keyword". Duolingo is "Duolingo: Language Lessons" and Strava is "Strava: Run, Bike, Walk". On iOS you get 30 characters for the name, 30 for the subtitle and 100 for a hidden keyword field.
- Design the first three screenshots as one ad. They appear in search results, where many people install without ever opening the page.
- Know your conversion rate. AppTweak measured an average of 8.56% from page view to download on the US App Store in 2025 and 16.15% on Google Play. If you are far below your category, fix the listing before sending it more traffic.
The full method is in the guides to App Store Optimization and keyword research.
Step 2: launch small and fix the first session
Before launch, set up three things. A one-page website with an email sign-up. A store pre-launch page: Apple takes pre-orders from 2 to 180 days before release, and Google Play has pre-registration. And a beta through TestFlight, which allows up to 10,000 external testers.
If you want Apple to consider featuring the launch, submit a featuring nomination in App Store Connect. Apple asks for at least two weeks' notice and recommends up to three months for wider featuring. It is a free form that most small developers never fill in.
Then release in one or two smaller countries before the big ones. A soft launch shows you whether people come back while the audience is small enough to fix things quietly.
What you are looking for is what happens on day zero. In RevenueCat's 2026 data on subscription apps, about half of all paid conversions happen on the day of install, and so do 55% of cancellations of three-day trials. People decide fast. If new users are not reaching the thing your app does within their first few minutes, no amount of marketing will make up for it. The guide to app engagement covers how to fix that.
Step 3: build the free channels
Store search
Review your keywords with every release and keep one screenshot test running. This is slow, compounding work, and it is the reason established apps pay less for growth than new ones.
Google search, your website and your app URLs
This is the channel app developers neglect most. People search Google for problems, and the stores never see those searches.
- Build pages that answer the problem, not pages that describe the app. For a budgeting app, "50/30/20 budget calculator" will bring more of the right visitors than a features page ever will.
- Write a page for every "[competitor] alternative" search in your category. Those visitors already want an app like yours and are unhappy with the one they have.
- Your App Store and Google Play URLs rank on Google too, mostly for your brand name. Apple states that your description "will be used for web engine search results", so write its first lines for a searcher even though it does nothing for App Store ranking.
- Add a Smart App Banner to your site. It is one meta tag, and it puts an install bar at the top of the page for every iPhone visitor using Safari.
- Set up universal links on iOS and app links on Android, so that visitors who already have the app are taken into it from your web pages.
Ratings
Ask for a rating right after the user succeeds at something. Apple allows the system prompt three times in 365 days, so never spend one on first launch. Reply to negative reviews. The reviewer is notified and can change the rating.
Creators
Small creators whose audience matches your users usually beat big general ones. Give each a tracked link or code so you know what they brought in. When a post performs, ask for permission to run it as an ad. TikTok and Meta both let you promote a creator's post from the creator's account, which keeps the look of a real recommendation instead of an advert.
Step 4: work out what an install is worth
This number sets your whole paid budget. Take your revenue over a period and divide it by the installs that produced it.
For comparison, RevenueCat's medians for subscription apps 60 days after install are $0.38 per install for apps with a free tier and $3.09 for apps with a hard paywall. A North American install is worth a median $0.55, and one in India or Southeast Asia $0.11.
Look at those figures before choosing where to advertise. If your install is worth 40 cents, you cannot profitably buy installs that cost several dollars each, and no bidding trick will change that. You either raise the value of an install, which is covered in the guide to app monetization, or you stay with free channels until you have.
Step 5: buy installs, one channel at a time
Apple Ads first
Search ads reach people who are already looking for an app like yours, so less of the budget goes on people with no interest. Apple's own recommended structure is four campaigns:
| Campaign | Keywords | Match type | Purpose |
|---|---|---|---|
| Brand | Your app and company name | Exact | Stops competitors taking your own searches |
| Category | What your app does | Exact | New users searching for a solution |
| Competitor | Similar apps | Exact | People comparing options |
| Discovery | Broad match, plus Search Match | Broad | Finds new search terms to move into the other three |
Start with brand and discovery. Brand is usually cheap and protects what you already earned. Discovery tells you which searches convert, and every converting term goes two places: into an exact-match campaign and into your organic keywords. Add each term you promote as a negative keyword in discovery so you do not bid against yourself.
Google App campaigns
App campaigns run across Search, Google Play, YouTube, Discover and the Display Network automatically. You give Google text, images, video and a bid. Begin by optimizing for installs. Once your purchase or trial events are flowing, switch the goal to that in-app action, because a campaign told to find installers will find the cheapest ones, not the ones who pay.
Meta and TikTok
Here the creative does the targeting. The platforms find your audience from who responds to the video, so the work is making videos. Plan on several new ones a week, shot on a phone, showing the app solving the problem in the first three seconds. Turn off what is not working within days and put the budget behind the one or two that are.
Measurement on iOS is limited. Since iOS 14.5 an app needs permission through App Tracking Transparency to track users across other companies' apps, and users are free to refuse, so expect campaign-level numbers rather than user-level ones.
That limit is one reason more subscription apps now send ad traffic to a web page first, take the sign-up and payment there, and only then send the user to the app. RevenueCat reports that 41% of the highest-earning apps it tracks now take some revenue on the web, against 1.3% of the smallest. If your paid social numbers do not work in-app, a web funnel is worth testing.
The numbers to watch
| Metric | What it is | Use it to decide |
|---|---|---|
| Conversion rate | Downloads divided by product page views | Whether the listing is ready for more traffic |
| Day 1, 7 and 30 retention | Share of new users who return | Whether the app is ready for paid installs |
| Revenue per install | Revenue divided by the installs that produced it | The most you can pay for a user |
| Cost per install (CPI) | Ad spend divided by installs from ads | Which channels are affordable |
| Return on ad spend (ROAS) | Revenue from ad-acquired users divided by ad spend | Which campaigns to scale and which to stop |
A plan for the first 90 days
- Weeks 1 to 2: keyword research, name, subtitle, keyword field, first three screenshots. Put up the landing page and open the beta. Submit a featuring nomination.
- Weeks 3 to 5: launch in one or two smaller markets. Watch how many new users reach the core action on day zero, and fix onboarding until most do.
- Weeks 5 to 7: add the rating prompt after a success moment. Start an Apple Ads brand campaign and a discovery campaign at a few dollars a day.
- Weeks 7 to 9: move converting search terms into your keywords and into exact-match campaigns. Publish the first two problem pages on your website.
- Weeks 9 to 11: calculate revenue per install. Test price and trial length if it is too low to advertise against.
- Weeks 11 to 13: launch in your main market. Scale the campaign whose cost per install sits below revenue per install, and add a second channel only when the first is profitable.
Mistakes that cost the most
- Buying installs before checking whether day-zero users reach the core action.
- Running Meta, TikTok, Google and Apple Ads at once on a small budget. None of them gets enough data to learn.
- Optimizing campaigns for installs long after you could optimize for purchases.
- Judging a channel on cost per install. A $1 install that never pays is more expensive than a $4 one that subscribes.
- Having no website. You give up every search that happens outside the stores.
FAQs
How do I market my app?
In this order: fix the store listing, launch in a small market and fix the first session, build organic installs from store search and Google, work out revenue per install, then buy installs on one channel at a time, starting with search ads.
How much does it cost to market an app?
The free channels cost time. Paid installs cost your price per install times the number you buy, so the real question is what you can afford to pay. That ceiling is your revenue per install, which for subscription apps has a 60-day median of $0.38 with a free tier and $3.09 with a hard paywall.
What is the difference between app marketing and ASO?
App Store Optimization is one part of app marketing. It covers ranking in store search and converting people who see the listing. App marketing also includes your website and SEO, paid advertising, creators, retention and monetization.
Do I need a website for my app?
Yes, if you want installs from Google. People search for problems on the web, and a page that answers one can rank and send visitors to the store. A website also lets you collect emails before launch and run paid traffic through a web funnel.
Can I market an app with no budget?
Yes. Store search, web pages built around the problem your app solves, a beta community and a featuring nomination all cost nothing but time. A few dollars a day on an Apple Ads discovery campaign is the first money worth spending, because it shows you which keywords convert.
Sources
- Apple Ads: App Store search statistics and campaign structure best practices
- RevenueCat: State of Subscription Apps 2026 (day-zero conversion, revenue per install and web revenue; medians for apps using RevenueCat)
- AppTweak: average app conversion rate per category, 2025
- Apple Developer: getting featured, TestFlight and platform version information
- App names are from the US App Store listings on October 5, 2026.